Fatty acids and their metabolism critically regulate podocyte survival

Fatty  Acids  and  their  Metabolism  Critically   Regulate  Podocyte  Survival  

Inauguraldissertation    

zur  

 Erlangung  der  Würde  eines  Doktors  der  Philosophie     vorgelegt  der  

 Philosophisch-­‐Naturwissenschaftlichen  Fakultät     der  Universität  Basel  

von  

Kapil  Dev  Kampe   aus  Hyderabad,  Indien  

Basel,  2014  

Genehmigt  von  der  Philosophisch-­‐Naturwissenschaftlichen  Fakultät   auf  Antrag  von  

Prof.  Dr.  Ed  Palmer   Prof.  Dr.  Marc  Donath   PD  Dr.  Andreas  Jehle    

Basel,  den  10.  Dezember  2013    

Prof.  Dr.  Jörg  Schibler   Dekan  der  Philosophisch-­‐

Naturwissenschaftlichen     Fakultät  

ABSTRACT

Diabetic nephropathy (DN) is the most common cause of end-stage renal disease in industrialized countries, and most affected patients have type 2 diabetes. Podocyte injury and loss are considered critical in the development, and progression of DN. Several factors of the diabetic milieu are well known to impair function and survival of podocytes. However, the role of free fatty acids (FFAs), which are elevated in type 2 diabetes, and the role of their metabolism are just emerging in the pathogenesis of DN. FFAs were reported to regulate podocyte survival. Saturated FFAs, i.e. palmitic acid, were found to induce endoplasmic reticulum (ER) stress and podocyte death, whereas monounsaturated FFAs, i.e. palmitoleic acid or oleic acid, were protective.

The aims of the present study were to investigate whether FFA metabolism is regulated in glomeruli of type 2 diabetic patients with DN and whether regulation of FFA metabolism affects the susceptibility of podocytes towards palmitic acid. Particularly, I aimed to investigate whether regulation of fatty acid oxidation (FAO) modifies palmitic acid-induced podocyte death. As genome wide association studies suggest that acetyl CoA carboxylase (ACC) 2, an important enzyme in the regulation of FAO, is involved in the pathogenesis of DN, I performed detailed studies investigating the role of ACCs in podocytes. Furthermore, I explored the effect of palmitic acid on podocytes in combination with well-known proapoptotic stimuli of the diabetic milieu.

The present study uncovered that palmitic acid can aggravate the toxicity of other factors which are known to be important in the pathogenesis of DN and which are considered to cause podocyte loss. In particular the toxicity of high glucose concentrations and transforming growth factor (TGF)-β are substantially increased by palmitic acid, whereas the effect of palmitic acid on tumor necrosis factor (TNF)-α induced podocyte death is discret.

In the main part of this study FFA metabolism and its effect on palmitic acid induced podocyte death was investigated. The study finds that in glomeruli of type 2 diabetic patients mRNA expression levels of several key enzymes involved in fatty acid metabolism are altered. Of particular relevance for my detailed studies on FAO, a significant upregulation of all three isoforms of carnitine palmitoyltransferase (CPT)-1, the rate-limiting enzyme for

FAO, and a downregulation of ACC-2, which catalyzes the formation of the CPT-1 inhibitor malonyl-CoA, are found which suggest a disposition for increased FAO. In vitro, stimulation of FAO by aminoimidazole-4-carboxamide-1β-D-ribofuranoside (Aicar) or by adiponectin, activators of the low-energy sensor AMP-activated protein kinase (AMPK), protect from palmitic acid induced podocyte death. Conversely, inhibition of CPT-1, a downstream target of AMPK, by etomoxir augments palmitic acid toxicity and impedes the protective Aicar effect. Etomoxir blocked the Aicar induced FAO measured with tritium labeled palmitic acid.

Of note, only double knockdown of ACC1 and ACC2 has a protective effect on palmitic acid induced cell death, which indicates that both isoforms contribute to the regulation of FAO in podocytes. Furthermore, the effect of Aicar is associated with a reduction of ER-stress as indicated by a significant attenuation of the palmitic acid induced upregulation of immunoglobulin heavy chain binding protein (BiP), an ER chaperone, and of the proapoptotic transcription factor C/EBP homologous protein (CHOP).

In conclusion, palmitic acid increases the toxicity of other factors known to contribute to podocyte loss, which underlines the potentially important contribution of elevated saturated FFAs in the pathogenesis of DN. An important role of FFAs and of their metabolism in the pathogenesis of DN is further suggested by profound changes in gene expression levels of key enzymes of FFA metabolism in glomerular extracts of type 2 diabetic patients. The changed expression profile indicates a compensatory, protective response. Moreover, the results of this study uncover that stimulation of FAO by modulating the AMPK-ACC-CPT-1 pathway protects from palmitic acid induced podocyte death. The results of this study should encourage further investigations to evaluate the therapeutic potential of interfering with FFA metabolism specifically with stimulating FAO for the prevention and therapy of DN.

TABLE OF CONTENTS

ABSTRACT ... I

LIST OF FIGURES AND TABLES ... VI

List of Figures ... VI List of Tables ... VII

List of ABBREVIATIONS ... VIII

1. Introduction ... 1

1.1. Diabetic Nephropathy (DN) ... 1

1.1.1. Incidence and prevalence of DN ... 1

1.1.2. Pathophysiology of DN ... 2

1.1.3. The role of podocytes in the pathogenesis of DN ... 2

1.1.4. Factors contributing to apoptosis of podocytes in DN ... 3

1.2. Lipotoxicity ... 4

1.2.1. Lipotoxicity in Diabetic Nephropathy ... 4

1.2.2. Lipotoxicity: The role of free fatty acids ... 5

1.2.2.1. The toxicity of saturated free fatty acids ... 5

1.2.2.2. Modulating pathways and the role of monounsaturated free fatty acids ... 8

1.2.3. Free fatty acids and their metabolism in podocytes ... 8

1.2.3.1. Regulation of podocyte survival by free fatty acids ... 8

1.2.3.2. Susceptibility of Podocytes to Palmitic Acid Is Regulated by Stearoyl-CoA Desaturases 1 and 2 …... ... 10

1.2.3.3. Lipotoxicity: Modulation by fatty acid oxidation and the role of ACCs ... 11

1.3. Aim of the study ... 13

2. MATERIALS and METHODS ... 14

2.1. Cell culture ... 14

2.2. Agonists, inhibitors and cytokines ... 15

2.3. Free fatty acids preparation ... 15

2.4. Apoptosis assay ... 15

2.5. ACC1 and ACC2 knock down ... 16

2.5.1. shRNA sequences and lentiviral expression vector ... 16

2.5.2. Lentiviral production ... 16

2.6. Western Blot ... 17

2.7. β-oxidation measurement ... 18

2.8. Statistical analysis ... 19

3. Results ... 20

3.1. Palmitic acid induced podocyte death: Modification by high glucose, TGF-β, and TNF α...20

3.1.1. Palmitic acid uncovers the toxicity of high glucose concentrations ... 20

3.1.2. TGF-β aggravates palmitic acid induced podocyte death ... 21

3.1.3. TNF alpha aggravates palmitic acid induced podocyte death ... 22

3.2. Regulation of fatty acid oxidation in palmitic acid induced podocyte cell death: Critical role of Acetyl CoA carboxylase 1 and 2 ... 23

3.2.1. Differential regulation of genes involved in fatty acid metabolism in glomeruli of patients with established DN ... 23

3.2.2. Modulation of fatty acid oxidation and its effect on palmitic acid induced podocyte death ...23

3.2.2.1. AMPK activation protects from palmitic acid induced cytotoxicity ... 23

3.2.2.2. Inhibition of AMPK exacerbates palmitic acid induced cell death and reversed the protection caused by Aicar. ... 26

3.2.2.3. Etomoxir aggravates palmitic acid induced podocyte death and reverses the protective Aicar effect ... 27

3.2.2.4. Aicar stimulates and etomoxir inhibits fatty acid oxidation ... 28

3.2.2.5. Combined silencing of ACC1 and ACC2 protects from palmitic acid induced podocyte death ... 29

3.2.2.6. Aicar reduces ER-stress and the upregulation of CHOP ... 32

4. Discussion ... 34

4.1. Aggravation of palmitic acid induced podocyte death by high glucose, TGF-β, and TNF-α ... 34

4.2. Regulation of fatty acid oxidation in palmitic acid induced podocyte death: critical role of acetyl CoA carboxylase 1 and 2 ... 35

5. Conclusion ... 38

6. References ... 40

7. Acknowledgements ... 49

7.1 Acknowledgements for micro-array data ... 50

APPENDIX ... a

LIST OF FIGURES AND TABLES

List of Figures

Figure 1: Prevalence of ESRD by primary diagnosis ... 1

Figure 2: Structure of the glomerular filtration barrier ... 2

Figure 3: ACC2 inhibits CPT-1 by the production of malonyl-CoA from acetyl-CoA ... 5

Figure 4: ER-stress and the unfolded protein response (UPR) ... 7

Figure 5: Palmitic acid induces apoptosis and necrosis of podocytes in a dose-dependent manner ... 9

Figure 6: Dose-and time-dependent induction of CHOP in podocytes by palmitic acid ... 9

Figure 7: Overexpressing SCD-1 partially protects from palmitic acid-induced apoptosis ... 10

Figure 8: Aicar stimulates fatty acid oxidation ... 11

Figure 9: High glucose accentuates the to toxicity caused by palmitic acid in podocytes ... 21

Figure 10: TGFβ increases the toxicity caused by palmitic acid in podocytes ... 21

Figure 11: TNFα induces podocyte death and aggravates palmitic acid induced toxicity ... 22

Figure 12: Differential expression of genes related to fatty acid metabolism in glomeruli of DN patients ... 23

Figure 13: Scheme representing the metabolic path activated by Aicar and Adiponectin ... 24

Figure 14: Aicar phosphorylates AMPK and ACC ... 24

Figure 15: Aicar attenuates palmitic acid induced podocyte death ... 25

Figure 16: Adiponectin decreases apoptosis induced by palmitic acid ... 25

Figure 18: Compound C accentuates palmitic acid induced podocyte death and partially reverses the protection by Aicar ... 27

Figure 19: Etomoxir exacerbates palmitic acid induced podocyte death ... 28

Figure 20: Etomoxir reverses the protection by Aicar for palmitc acid induced podocyte death ... 28

Figure 22: Etomoxir prevents the Aicar induced β-oxidation ... 29

Figure 23: Immunoblot for ACC1, ACC2 and ACC1/2 knock down ... 30

Figure 24: ACC1 or ACC2 single knockdown is not protective for palmtitic acid induced podocyte death ... 30

Figure 25: Combined knock down of ACC1 and ACC2 protects podocytes from palmitic acid induced cell death ... 31

Figure 26: Combined knock down of ACC1 and ACC2 with a second set of shRNAs against ACC1 and ACC2 protects podocytes from palmitic acid induced cell death ... 31

Figure 27: Modest effect of Aicar in ACC1/ACC2 double silenced podocytes ... 32

Figure 28: Aicar mitigates palmitic acid induced ER-stress ... 33

Figure 29: A working model for the Aicar activated AMPK-ACC-CPT-1 pathway and the prosurvival effects of oleic acid and Scd-1/-2 on palmitic acid-induced podocyte death. ... 39

List of Tables

Table 2: shRNA sequences and respective plasmids. ... 16

Table 3: Summary of transfection conditions for the production of lentiviral particles. ... 17

Table 4: List of primary and secondary antibodies for western blot. ... 18

LIST OF ABBREVIATIONS

ACC Acetyl-CoA carboxylase

Aicar 5-aminoimidazole-4-carboxyamide ribonucleoside

AMPK AMP-activated protein kinase

ATF Activating transcription factor

ATP Adenosine triphosphate

BiP Immunoglobulin heavy chain binding protein

BSA Bovine serum albumin

CHOP C/EBP homologous protein

CPT-1 Carnitine palmitoyl transferase 1

DAG Diglyceride

DGAT1 DAG acyltransferase 1

DMEM Dulbecco’s modified eagle medium

DMSO Dimethyl sulfoxide

DN Diabetic nephropathy

DNA Desoxyribonucleic acid

dNTP Deoxyribonucleoside triphosphate

DPM Disintegrations per minute

EDTA Ethylenediaminetetraacetic acid

EGTA Ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′- tetraacetic acid

ER Endoplasmic reticulum

ERAD ER associated protein degradation

ESRD End-stage renal disease

FBS Fetal bovine serum

FFA Free fatty acid

GBM Glomerular basement membrane

HEK Human embryonic kidney

IRE-1 Inositol-requiring enzyme 1

IRS Insulin receptor substrate

JNK c-Jun NH2-terminal kinase

LXR Liver X receptor

MUFA Monounsaturated fatty acid

PBS Phosphate-buffered saline

PCR Polymerase chain reaction

PERK Double-stranded DNA-dependent protein kinase

(PKR)-like ER kinase

PI Propidium iodide

PPARα Peroxisome proliferator-activated receptor α

RIPA Radioimmunoprecipitation assay

ROS Reactive oxygen species

RPM Revolutions per minutes

RPMI Roswell Park Memorial Institute

SCD Stearoyl-CoA desaturase

SD Standard deviation

SDS-PAGE Sodium dodecyl sulfate polyacrylamid gel electrophoresis

SFA Saturated fatty acid

SNP Single nucleotide polymorphism

TBS Tris-buffered saline

TG Triglyceride

TGF-ß Transforming growth factor ß

TO TO901317

UPR Unfolded protein response

VSV Vesicular stomatitis virus

1. INTRODUCTION

In the following I will give an introduction into diabetic nephropathy (DN). Thereby, the main focus will lie on podocytes which are highly specialized epithelial cells of the glomerular filtration barrier and which are thought to be critically involved in the pathogenesis of DN. In the second part I will introduce the term “lipotoxicity” and its potential relevance in the pathogenesis of DN. As saturated and monounsaturated free fatty acids critically determine lipotoxicity, I will review the literature with a special focus on their cellular effects and the mechanisms involved herein. Lastly, I will summarize two recent studies related to free fatty acids (FFAs) and FFAs metabolism in podocytes in which I have been involved as a coauthor and which are directly linked to my main PhD thesis project on fatty acid oxidation and acetyl CoA carboxylases (ACCs) in podocytes.

1.1. Diabetic Nephropathy

Diabetic nephropathy (DN) is the most common cause of end-stage renal disease (ESRD) in industrialized countries, e.g. over 40% in the US (Figure 1) (USRDS Annual report, (2013)).

The majority of diabetic patients starting renal replacement therapy today have type II diabetes as the prevalence of type II diabetes is much higher. Of the patients with type II diabetes 20-40% develop ESRD (Foley et al., 1998). The five-year survival rate of patients with DN and renal replacement therapy is significantly worse than in patients with other renal diseases mainly as a result of an increased cardiovascular mortality (Locatelli et al., 2004;

USRDS, 2011). Therefore, it is important to better understand the pathogenesis of DN, to identify new strategies and additional therapeutic targets for the prevention and treatment of DN.

Figure 1: Prevalence of ESRD by primary diagnosis (U.S. Renal Data System, USRDS 2013 Annual Data Report, Chapter One, Figure 1-15, Volume Two)

1.1.2. Pathophysiology of DN

Dating back to the first description by Kimmelstiel and Wilson (Kimmelstiel P, 1936) the typical lesion of DN is mesangial matrix expansion accompanied by hypertrophy of mesangial cells, and thickening of the glomerular basement membrane (GBM). In addition, already in these seminal reports intraglomerular lipid deposits were described.

The classical, earliest clinical sign of DN is microalbuminuria, i. e. loss of small amount of albumin in the urine, resulting from damage to the glomerular filtration barrier.

Consequently, attention to mechanisms focusing on alterations of the glomerular filtration barrier seems most warranted.

The glomerular filtration barrier is made of three interdependent layers (Figure 2), which are fenestrated endothelial cells, the GBM, and very specialized epithelial cells, the so called podocytes (Figure 2). All these layers form a size- and charge selective renal filtration sieve to prevent albumin and other molecules to be lost in the urine.

Figure 2: Structure of the glomerular filtration barrier. A) Schematic picture of the glomerular filtration barrier consisting of fenestrated endothelium cells, the glomerular basement membrane (GBM), and podocytes with their interdigitating foot process (Image from: J Patrakkaa and K Tryggvasona. Biochem Biophys Res Comm., 2010 (Patrakka and Tryggvason, 2010)); B) Image taken using a scanning electron microscope of a podocyte wrapped around a glomerular capillary (Image from: Smoyer WE & Mundel P, J Mol Med, 1998 (Smoyer and Mundel, 1998)).

1.1.3. The role of podocytes in the pathogenesis of DN  

Over the past two decades by elucidating the genetic origin of a number of single human gene defects that result in congenital or early onset focal segmental glomerulosclerosis with massive proteinuria, it has become apparent that podocytes are the primary affected cell and critically determine the proper function of the glomerular filtration barrier (Kriz, 2003; Reidy and Kaskel, 2007). Also, in many other renal diseases with proteinuria including DN

A   B  

increasing evidence suggests that podocyte dysfunction and loss mainly contribute to the development of proteinuria (Jefferson et al., 2008; Shankland, 2006). Podocytes are pericyte- like cells. They have a complex cellular architecture that includes a cell body, primary processes that further ramify into fine secondary foot processes. Individual foot processes interdigitate with foot processes of neighboring cells, and the filtration slits between the processes are bridged by slit diaphragms, which critically contribute to the selective permeability of the glomerular filtration barrier (Mundel and Kriz, 1995; Reiser et al., 2000).

Moreover, podocytes critically determine the biophysical characteristics of the GBM and they are important to counteract the hydrostatic pressure from the glomerular capillaries (Endlich and Endlich, 2006).

Podocytopathy in DN is characterized by foot process widening. Importanly, this morphological change correlates in type I diabetic subjects with the urinary albumin excretion rate (Berg et al., 1998). Both, in patients with type I or type II diabetes the number and density of podocytes have been reported to be decreased (Dalla Vestra et al., 2003;

Pagtalunan et al., 1997; Steffes et al., 2001; White and Bilous, 2000; White et al., 2002), and this podocyte loss relates to proteinuria (White and Bilous, 2004). A study performed in Pima Indians with type II diabetes suggests that a reduced number of podocytes per glomerulus predicts progressive kidney disease (Meyer et al., 1999).

Podocytes have no or very limited ability to replicate (Marshall and Shankland, 2006), therefore podocyte death and/or podocyte detachment from the GBM are thought to account for podocyte loss. Indeed, podocyturia has been documented in patients with type II diabetes and seems to correlate with disease progression (Nakamura et al., 2000).

1.1.4. Factors contributing to apoptosis of podocytes in DN  

It is likely that multiple hits are necessary for the occurrence of injury and ultimately apoptosis in podocytes during the development of diabetic nephropathy.

In vitro, high glucose levels induce apoptosis in podocytes, and increased ROS have been shown to be important mediators of glucotoxicity (Susztak et al., 2006). In db/db mice chronic inhibition of NADPH oxidase was able to reduce podocyte apoptosis and ameliorated podocyte depletion, urinary albumin excretion, and mesangial expansion (Susztak et al., 2006). Of note, increased ROS levels in diabetes do not result from hyperglycemia alone, but

angiotensin II (Haugen et al., 2000) as well as elevated free fatty acids (Piro et al., 2002)) may also be important contributors.

TGF-β1 mRNA and protein levels are increased in various models of diabetes in rodents, and TGF-β signaling can be activated by a large number of mediators in diabetes including ROS, angiotensin II, and advanced glycation products (Ziyadeh, 2004). In vitro, TGF-β has been reported to induce apoptosis in murine podocyte (Schiffer et al., 2001). In podocytes derived from glomeruli of rats angiotensin II was shown to have a pro-apoptotic effect also (Ding et al., 2002). Interestingly, this effect was shown to depend on angiotensin II induced TGF-β production and could be attenuated by anti- TGF-β antibodies. Most importantly, in db/db mice administration of a neutralizing anti-TGF-β antibody was found to prevent the mesangial matrix expansion and to protect from a decline in kidney function (Ziyadeh et al., 2000). Also, tumor necrosis factor (TNF)-α was reported to induce podocyte death (Ryu et al., 2012; Tejada et al., 2008), and a variety of direct and indirect evidence suggests that TNF- α plays an important role in the pathogenesis of DN (Navarro-Gonzalez et al., 2009).

1.2. Lipotoxicity

Obesity, the metabolic syndrome, and type 2 diabetes are associated with elevated serum triglycerides and free fatty acids (FFAs). This leads to lipid accumulation in nonadipose tissues, including pancreas, heart, liver, and kidneys. Accumulation of excess lipids in these organs causes cell dysfunction and cell death. This process is termed lipotoxicity (Brookheart et al., 2009).

1.2.1. Lipotoxicity in Diabetic Nephropathy  

In the kidneys of diabetic humans, intraglomerular lipid deposits were described first in 1936 by Kimmelstiel and Wilson and subsequently observed by other researchers (Kimmelstiel and Wilson, 1936; Lee et al., 1991). Upregulated lipogenic genes and development of glomerular and tubular lipid deposits have been observed in different animal models of obesity and diabetes mellitus (mice fed high-fat diets, leptin impaired db/db and ob/ob mice, streptozotocin-treated rats) (Jiang et al., 2007; Kume et al., 2007; Sun et al., 2002; Wang et al., 2005).

Importantly, recent data from genome wide association studies (GWAS) suggest a potentially important role of FFA metabolism in the pathogenesis of DN. In detail, two GWAS in type 2 diabetic patients found a polymorphism in a noncoding region of acetyl-CoA carboxylase (ACC) 2 with a strong association with proteinuria (Maeda et al., 2010; Tang et al., 2010).

ACCs catalyze the carboxylation of acetyl-CoA to produce malonyl-CoA, which inhibits CPT-1, the rate limiting enzyme of FAO (Figure 3). As the DN-risk SNP of ACC2 results in a higher ACC2 expression (Maeda et al., 2010), it can be postulated that this leads to increase in malonyl-CoA levels and increased inhibition of CPT-1 with subsequent impairment of FAO.

Figure 3: ACC2 inhibits CPT-1 by the production of malonyl-CoA from acetyl-CoA

1.2.2. Lipotoxicity: The role of free fatty acids  

Elevated FFAs and disturbed FFA metabolism are critical determinants of lipotoxicity (Brookheart et al., 2009). Toxicity has been attributed mainly to saturated fatty acids (SFAs) whereas monounsaturated fatty acids (MUFAs) exert beneficial and cytoprotective effects (Brookheart et al., 2009; Nolan and Larter, 2009). Up to 80% of the plasma FFAs consist of the saturated palmitic (C16:0) and stearic acid (C18:0) as well as the monounsaturated oleic acid (C18:1) (Hagenfeldt et al., 1972).

1.2.2.1. The  toxicity  of  saturated  free  fatty  acids    

The toxicity of SFAs has been attributed to multiple cellular mechanisms. One mechanism is related to decreased triglyceride (TG) synthesis and accumulation of cytotoxic metabolites such as diacylglycerides (DAGs). Of note, MUFAs can induce fatty acid oxidation and increase lipid storage in form of TGs, thereby reducing cytotoxic metabolites such as DAGs (Nolan and Larter, 2009), and DAG-mediated lipotoxicity may depend on the saturation of

fatty acids incorporated in DAGs (Bergman et al.), i. e. DAG with a higher content of MUFAs are less toxic. A second mechanism is related to ceramide synthesis. Palmitic acid is a substrate for the production of ceramide (Listenberger and Schaffer, 2002). Ceramide is a lipid secondary messenger involved in initiation of apoptosis. Mechanistically, ceramide induces insulin resistance and thereby may affect the prosurvival effect of insulin signaling;

also, ceramide induces apoptotic pathways via increased membrane permeability of mitochondria (Bikman and Summers, 2011; Siskind, 2005). Inhibition of ceramide synthesis prevents lipotoxicity in pancreatic β-cells but not fibroblasts, suggesting that cell type-specific metabolic channeling of FFAs may be important (Brookheart et al., 2009). In cardiomyocytes, palmitic acid leads to depletion of cardiolipin, a phospholipid localized to the inner mitochondrial membrane and important for optimal mitochondrial function (Chicco and Sparagna, 2007), and reduced cardiolipin levels are thought to contribute to disruption of the mitochondrial inner membrane with release of cytochrome c. (Ostrander et al., 2001). SFAs as palmitic acid also effect other mitochondrial membrane phospholipids and thereby disturb mitochondrial function and lead to increase production of ROS (Brookheart et al., 2009).

Importantly, oxidative stress can impair membrane integrity, organelle function, and gene expression, and thereby contributing to cell death. A third mechanism linked to the toxicity of SFA is related to the endoplasmic reticulum (ER). Palmitic acid rapidly increases the saturated lipid content of the ER leading to compromised ER morphology and integrity (Borradaile et al., 2006). In pancreatic β-cells palmitic acid depletes ER Ca²⁺ and slows ER Ca²⁺ uptake (Cunha et al., 2008). Disturbed ER homeostasis with the accumulation of mis- or un-folded proteins referred to as ER-stress (Rasheva and Domingos, 2009). ER-stress results in several signaling pathways, collectively known as unfolded protein response (UPR). The UPR cascade involves three signaling branches that are mediated by ER transmembrane receptors: double-stranded DNA-dependent protein kinase (PKR)-like ER kinase (PERK), inositol-requiring enzyme 1 (IRE-1), and activating transcription factor 6 (ATF6) (Figure 4).

These receptors are bound by the ER chaperone immunoglobulin heavy chain binding protein (BiP, also termed GRP78 or HSPA5) that keeps them silenced. Accumulation of unfolded and/or misfolded proteins is leading to dissociation of BiP and therefore to the activation of PERK, IRE-1 and ATF6. The UPR is primarily an adaptive response to maintain and restore proper ER function (Kaufman, 2002; Ma and Hendershot, 2001). However, the UPR is also linked to inflammatory signals (Figure 4), which themselves can trigger or maintain ER- stress. In addition, ER-stress leads to insulin resistance, at least in part through serine phosphorylation of insulin receptor substrate 1 (IRS1) by IRE1-activated JNK1 (Figure. 4)

(Ozcan et al., 2004). Unresolved and severe ER-stress may lead to apoptosis through up- regulation of the proapoptotic transcription factor CHOP (also known as DDIT3) (Zinszner et al., 1998), and loss of CHOP protects β cells from apoptosis in the db/db mouse model (Song et al., 2008). Conversely, overexpression of BiP in pancreatic beta-cells can reduce palmitic acid induced apoptosis (Laybutt et al., 2007) which may be explained by the ability of BiP to bind to and thereby repress the activity of the ER-stress transducers (Bertolotti et al., 2000) (Figure 4).

Figure 4: ER-stress and the unfolded protein response (UPR). In eukaryotic cells monitoring of the ER lumen is mediated by three ER membrane-associated proteins , PERK (PKR-like eukaryotic initiation factor 2α kinase), IRE1 (inositol requiring enzyme 1), and ATF6 (activating transcription factor-6). In a well-functioning and

“stress-free” ER, these three transmembrane proteins are bound by a chaperone, BiP/GRP78, in their intralumenal domains and rendered inactive. Accumulation of improperly folded proteins and increased protein cargo in the ER results in the recruitment of BiP away from these UPR sensors. This results in oligomerization and activation of the two kinases, PERK and IRE1. Activation of the third branch of the UPR requires translocation of ATF6 to the Golgi apparatus where it is processed to an active transcription factor. The endoribonuclease activity of IRE1α cleaves the mRNA of the X-box binding protein-1 (XBP1), creating an active (spliced) form of the transcription factor (XBP1s). XBP1s, alone or in conjunction with ATF6α, launches a transcriptional program to produce chaperones (e.g. BiP) and proteins involved in ER-associated protein degradation (ERAD). ATF6 regulates XBP1 mRNA expression, and in addition ATF6 regulates together with XBP1s the expression of ER chaperones and UPR quality control genes. PERK activation results in phosphorylation of eIF2α (eukaryotic translational initiation factor 2α), which converts eIF2α to a competitor of eIF2B resulting in reduced global protein synthesis and a subsequent reduction in the workload of the ER. All three branches of the UPR are involved in the regulation/activation of the NF-κB-IKK pathway leading to an inflammatory response. If the ER-stress is prolonged, the UPR can induce apoptosis involving PERK mediated

activation of ATF4, which induces genes involved in apoptosis, e.g. C/EBP homologous protein (CHOP).

(Adapted from (Hotamisligil, 2010))

1.2.2.2. Modulating pathways and the role of monounsaturated free fatty acids

In contrast to SFAs, monounsaturated fatty acids (MUFAs) exert mainly cytoprotective effects (Brookheart et al., 2009; Nolan and Larter, 2009). MUFAs are more potent ligands of the peroxisome proliferator-activated receptor α (PPARα), a transcription factor regulating lipid metabolism (Keller et al., 1993). PPARα is inducing transcription of genes involved in mitochondrial ß-oxidation (Hihi et al., 2002), and increasing ß-oxidation is thought to be a mechanism to detoxify cells from SFAs. In addition, MUFAs favor the incorporation of SFAs in TG, and this is thought to be cytoprotective as palmitic acid and its metabolites incorporated in TG are thought to be “biologicaly inert”, i. e. they are stored away in “safe lipid pools” (Nolan and Larter, 2009). Mechanistically, incorporation of SFA-derived acyl- CoAs has been suggested to be more efficient in the presence of MUFAs as MUFAs are the preferred substrates for acyl-CoA:diacylglycerolacyltransferases (DGATs), that transfer acyl- CoAs to DAGs to form TGs (Cases et al., 1998; Cases et al., 2001; Hardy et al., 2003; Ricchi et al., 2009). Similarly, overexpression of stearoyl-CoA desaturases (SCDs) that desaturate saturated FFAs to form MUFAs has been shown to result in resistance to palmitic acid induced cell death (Listenberger et al., 2003). Furthermore and as previously discussed, ER- stress plays an important role in lipotoxicity. Importantly, MUFAs are directly linked to attenuation of ER-stress (Diakogiannaki et al., 2008; Holzer et al., 2011).

1.2.3. Free fatty acids and their metabolism in podocytes  

In the following I will summarize recent studies, which are closely linked, to my own PhD project and in which I was able to contribute as a coauthor. Also, I will introduce in more detail the role of fatty acid oxidation and its regulation by ACCs.

1.2.3.1. Regulation of podocyte survival by free fatty acids  

Only recently the effect FFAs has been investigated in podocytes. For these studies our laboratory used conditionally immortalized mouse podocytes, a well-established model to study podocyte biology in vitro (Mundel et al., 1997). As shown in figure 5 palmitic acid dose dependently increased both apoptosis and necrosis in podocytes. Podocyte cell death was assessed by flow cytometry after staining with Annexin V and propidium iodide (PI).

Annexin V-positive/PI-negative podocytes were considered apoptotic, whereas annexin V- positive/PI-positive podocytes were considered (late apoptotic) necrotic cells .

Figure 5: Palmitic acid induces apoptosis and necrosis of podocytes in a dose-dependent manner. Podocytes

were exposed to palmitic acid (125 – 500 µM) or BSA (at a concentration quivalent to cells treated with 500 µM palmitic acid complexed to BSA) for 38 h. Quantitative analysis of palmitic acid induced podocyte cell death.

Bar graph represents the mean percentages +/- SD of annexin V-positive/PI-negative (early apoptotic) or annexin V-positive/PI-positive (late apoptotic/necrotic) podocytes (n=3; * p<0.05, ** p < 0.01).(Sieber et al., 2010)

Similar to other cell type (Guo et al., 2007; Kharroubi et al., 2004; Martinez et al., 2008; Wei et al., 2009; Wei et al., 2006) we observed that palmitic acid dose- and time-dependently results in ER-stress as indicated by the induction of the proapoptotic transcription factor CHOP, which is typically upregulated during severe ER-stress (Zinszner et al., 1998) (figure 6), (Sieber et al., 2010).

Figure 6: Dose-and time-dependent induction of CHOP in podocytes by palmitic acid (

Importantly, further experiments demonstrated that palmitic acid induced podocyte death as well as ER-stress can be prevented by coincubation with palmitoleic or oleic acid (Sieber et al., 2010). Finally, we were able to demonstrate that palmitic acid induced cell death at least in part is mediated by CHOP as knockdown of CHOP led to a significantly reduction of podocyte death (Sieber et al., 2010). In addition, we and other observed that palmitic acid

leads to insulin resistance in podocytes (Sieber and Kampe, unpublished observation, and (Lennon et al., 2009)). As insulin signaling has a strong prosurvival effect in many cell types (Huber et al., 2003) this may further increase the susceptibility of podocyte to proapoptotic stimuli such as palmitic acid.

1.2.3.2. Susceptibility of Podocytes to Palmitic Acid Is Regulated by Stearoyl-CoA Desaturases 1 and 2

To study the impact of type 2 diabetes and diabetic nephropathy (DN) on glomerular fatty acid metabolism microarray analysis of key enzymes involved in fatty acid metabolism was performed. Specifically, our laboratory investigated the gene expression in glomeruli of patients with type 2 diabetes mellitus compared to pretransplantation living donors (Sieber et al., 2013). Interestingly, the most prominent change was the upregulation of stearoyl-CoA desaturases (SCD)-1, and by immunohistochemistry the increased signal for SCD-1 could be predominantly observed in podocytes (Sieber et al., 2013). To address the potential contributory role of FFAs to the altered gene expression profile in glomeruli and podocytes of patients with DN, cultured podocytes were treated with 200µM palmitic acid complexed to BSA, compared to uncomplexed BSA, and significantly increased expression of Scd-1 (1.7 ± 0.7-fold) and Scd-2 (1.9 ± 0.6-fold), the most abundant SCD isoforms in murine kidneys (Ntambi and Miyazaki, 2003) and murine podocytes was observed. Further experiments with pharmacological and genetic overexpression of SCDs as well as gene silencing of SCD-1, and -2 elegantly demonstrated that both isoforms are protective for palmitic acid induced podocyte death Figure 7, (Sieber et al., 2013).

Figure 7: Overexpressing SCD-1 partially protects from palmitic acid-induced apoptosis. Podocytes with SCD- 1 overexpressing were compared to podocytes with overexpression of GFP. SCD-1 reduced palmitic acid- induced apoptosis and necrosis in podocytes compared to GFP controls. Bar graph shows mean percentages ± SD of apoptotic and necrotic cells after exposure to 200µM palmitic acid for 48 h (n = 3, * p < 0.05, ** p < 0.01, (Sieber et al., 2010)

Mechanistically, MUFAs or stimulation of SCDs, which convert saturated FFAs to MUFAs, promote the incorporation of palmitic acid into TG, suggesting that the protective effect at least in part results from compartmentalization of palmitic acid in “safe lipid pools” (Sieber et al., 2013).

1.2.3.3. Lipotoxicity: Modulation by fatty acid oxidation and the role of ACCs

Studies in endothelial cells suggest that stimulation of fatty acid oxidation (FAO) protects from palmitic acid induced cell death (Borradaile et al., 2006). In these studies FAO was stimulated by the by the AMPK agonist aminoimidazole-4-carboxamide-1β-D-ribofuranoside (Aicar) which leads to phosphorylation of ACCs resulting in lower malonyl-CoA levels and disinhibtion of CPT-1 (Fig. 9), the rate-limiting enzyme of FAO (Muoio and Newgard, 2008).

Figure 8: Aicar stimulates fatty acid oxidation. Figure depicts the metabolic pathway activated by Aicar, which results in stimulation of fatty acid oxidation.

As mentioned above (section 1.2.1.), two recent GWAS in type 2 diabetic patients found a polymorphism in a noncoding region of ACC2 with a strong association with proteinuria (Maeda et al., 2010; Tang et al., 2010). The DN-risk single nucleotide polymorphism of ACC2 results in a higher ACC2 expression (Maeda et al., 2010) potentially leading to increased malonyl-CoA levels and decreased FAO.

In humans and rodents there are two ACC isoforms, ACC1 (ACC alpha) and ACC2 (ACC beta) (Savage et al., 2006), which share considerable sequence identity and the same domain structure responsible for enzyme activity (Savage et al., 2006). In contrast to ACC1, ACC2 has an extra N-terminal hydrophobic domain, which facilitates its localization to the

mitochondrial membrane (Abu-Elheiga et al., 2005), where it preferentially regulates local malonyl-CoA levels and CPT-1 activity. In contrast cytosolic ACC1 is classically thought to regulate malonyl-CoA synthesis for incorporation into fatty acids in lipogenic tissues.

However, more recently this classical view has been challenged, and at least in some cell types, e.g. hepatocytes, both isoforms have been shown to regulate CPT-1 activity synergistically (Savage et al., 2006).

1.3. Aim of the study

Increasing evidence suggests that damage and loss of podocytes are early events in DN and critically determine disease progression. Several factors of the diabetic milieu are known to impair function and survival of podocytes. Although lipid accumulation is a well known feature of DN, only recently the potentially important role of FFAs and FFA metabolism in this process were acknowledged.

Over the last five years our laboratory systematically analyzed the effects of FFAs as well as FFA metabolism in podocytes. A key finding was that FFAs can regulate podocyte survival.

Specifically, SFAs, i.e. palmitic acid, were found to induce endoplasmic reticulum (ER) stress and podocyte death, whereas monounsaturated FFAs, i.e. palmitoleic acid or oleic acid, were protective.

The aims of the present study were to investigate whether FFA metabolism is regulated in glomeruli of type 2 diabetic patients with DN and whether regulation of their metabolism affects the susceptibility of podocyte towards palmitic acid. The main focus was to understand whether regulation of FAO modifies palmitic acid-induced podocyte death. As genome wide association studies suggest that acetyl CoA carboxylase (ACC) 2, an important enzyme in the regulation of FAO, is involved in the pathogenesis of DN, detailed studies investigated the role of ACCs in podocytes. Furthermore, I investigated whether palmitic acid modifies podocyte death induced by other factors of the diabetic milieu, which have been shown to be involved in the pathogenesis of DN and which are thought to contribute to damage and loss of podocytes.

2. MATERIALS AND METHODS

2.1. Cell culture

Podocytes were cultured following the protocol described by Mundel et al (Shankland et al., 2007). Conditionally immortalized mouse podocyte cell lines were established from the immortomouse, which carries a thermosensitive (ts58A) variant of the SV 40 T antigen as a transgene (Shankland et al., 2007). Podocytes having the passage number from 4 to 14 are utilized for performing the experiments. First, podocytes were cultured or proliferated in permissive conditions which include growing in 33°C with 50U/ml interferon gamma (IFN-γ,

# CTK-358-2PS, MoBiTec GmbH, Germany) for first two passages, later IFN-γ concentration can be brought down to 10U/ml. Differentiation of podocytes is done in non-permissive conditions which includes thermoshift to 37°C without IFN-γ. Podocytes were allowed to undergo differentiation atleast 11 days prior to the start of experiments.

Podocytes are cultured in RPMI-1640 (#21875, Invitrogen) supplemented with 10% FBS (#10270, Invitrogen), 100 U/ml penicillin and 100 µg/ml streptomycin (#15140, Invitrogen).

For apoptosis experiments, β-oxidation experiments 6-well plates were employed and for isolating mRNA and protein 10-cm dishes were employed from BD biosciences. All the plates and dishes were coated with 0.1 mg/ml type I collagen (BD biosciences) prior to seeding the cells. Freezing of the cells was performed in complete culture medium supplemented with 8% (v/v) dimethylsulfoxide (Sigma). For the production of lentiviral particles, HEK293 cells were employed as packaging cells. HEK293 cells were cultured in DMEM (#41965, Invitrogen) supplemented with 10% FBS and penicillin/streptomycin.

2.2. Agonists, inhibitors and cytokines

Substance Supplier (catalog number) Physiological role Concentration applied Aicar Cell signaling (#9944) AMPK agonist 0.5mM

Adiponectin BioVision (#4902-100) AMPK agonist 0.5µg/ml

Compound C Sigma (#5499) AMPK inhibitor 4µM

Etomoxir Sigma (#E1905) CPT-1 inhibitor 10, 30 and 200µM

TGFβ Roche (#10874800) 5ng/ml

TNFα Sigma (#T7539) 5ng/ml

Table 1: Agonists, inhibitors and cytokines

2.3. Free fatty acids preparation

Sodium palmitate and oleic acid (both from Sigma) were dissolved overnight at 10 mM in glucose-free RPMI-1640 medium (#11879) containing 11% essential fatty-acid free BSA (Sigma) under N2-atmosphere at 55°C, sonicated for 10 min and sterile filtered (stock solution). The molar ratio of fatty acid to BSA was 6:1.The effective free fatty acid concentrations were measured with a commercially available kit (Wako). Endotoxin concentration was equal or less than 0.5ng/ml, as determined by a kit (#L00350) from Genscript (Piscataway, NJ, USA).

2.4. Apoptosis assay

The cells were trypsinized, washed once with PBS, and resuspended in 120 µl annexin V binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl2, pH 7.4). 100 µl of the cell suspension was used for the staining procedure. Alexa-647 annexin V (#A23204, Invitrogen) staining was applied for 15 min at room temperature at a dilution of 1:100 (see producer protocol) and before analyzing an additional 400 µl of annexin V binding buffer was added along with 0.5µg propidium iodide (#P3566, Invitrogen) were added. 20’000 – 25’000 cells were analyzed by flow cytometery with CyAn™ ADP Analyzer (Beckman Coulter). Data from flow cytometry was analyzed by FLOWJO (Tree Star, Inc. Ashland, OR, USA) software program. Annexin V-positive/PI-negative podocytes were considered apoptotic, whereas annexin V-positive/PI-positive podocytes were considered (late apoptotic) necrotic cells.

2.5. ACC1 and ACC2 knock down

2.5.1. shRNA sequences and lentiviral expression vector

For the knock down, ACC1, ACC2 and scrambled shRNA sequences were cloned into pSIH- H1-puro lentiviral expression plasmid, which is a kind gift of Dr. Markus Heim, (University Hospital Basel, Switzerland). Respective details of shRNA sequences are furnished in table 1 along with information of plasmids and references. pSIH-H1-puro was first linearized with EcoR1/BamH1 restriction enzymes and at the same time shRNA oligos of respective genes were designed and ordered with restriction sites for EcoR1/BamH1 flanking them to facilitate cloning into the pSIH-H1-puro vector. Next, shRNA oligos were ligated with digested pSIH- H1-puro and ligation product was transformed into competent E.coli strain, DH5-α.

Following day of bacterial transformation, colonies were randomly picked and miniprep was done to isolate plasmid and which further sent to sequencing for confirmation.

Gene shRNA sequence (5’ – 3’) Vector GeneBank

number

Reference

ACC1 GCAGATTGCCAACATCCTAGA pSIH-H1-puro NM_133360 (Jeon et al., 2012) ACC2 GTGGTGACGGGACGAGCAA pSIH-H1-puro NM_133904 (Jeon et al., 2012)

ACC2 (2) GAGGTTCCAGATGCTAATG pSIH-H1-puro NM_133904 Optimized

Scrambled GACCGCGACTCGCCGTCTGCG pSIH-H1-puro (Sieber et al., 2010) Table 2: shRNA sequences and respective plasmids.

2.5.2. Lentiviral production  

A 4-plasmid based lentiviral system (kindly provided by Dr. Markus Heim, University Hospital Basel, Switzerland) was employed with following helper plasmids: pRSV-REV (Rev expression vector), pMDLg/pRRE (Gag-Pol expression vector), and pMD2.G (VSV-G expression vector). All helper plasmids along with pSIH-H1-puro were mixed with 45µl of FuGene HD (Promega, Madison, WI, USA) transfection agent in total 3ml of Opti-MEM (#31985, Gibco) and incubated at room temperature for 20min. Next HEK293 were transfected in a 10-cm dish with 5ml of DMEM (#41965) without antibiotics and medium was changed after 8-12h, followed by addition of fresh complete 10ml of DMEM (conditions for transfection is detailed in table 2). HEK293 cells were grown upto 60 to 70% confluence

before transfection. 48h post transfection, the supernatant medium enriched with lentiviral particles was harvested, spun at 780g for 5min and filtered through 0.45µm filter.

Transduction of podocytes was done by adding virus containing media with pre-treating podocytes with 5µg/ml polybrene (Sigma) for 5min.  8-24 hours after transduction, medium was exchanged. Experiments were performed three to five days after viral transduction.

Amount of viral particles to be used for transducing cells was standardized separately by employing a GFP-based lentiviral expression plasmid FUGW, which facilitates the visual inspection of the efficiency of viral transduction. For all the experiments a viral titer to achieve efficiency of 70 – 80% was employed.

Plasmid/Reagent Amount

pSIH-H1-puro (with respective shRNA sequence) 9 µg Rev expression vector (pRSV-REV) 1.8 µg Gag-Pol expression vector (pMDLg/pRRE) 4.5 µg VSV-G expression vector (pMD2.G) 2.7 µg

Total plasmid DNA 18 µg

FuGene HD 45 µl

Total Opti-MEM 3 ml

Table 3: Summary of transfection conditions for the production of lentiviral particles.

2.6. Western Blot

For protein isolation cells were always cultured in 10-cm dishes. For isolating protein, medium was sucked off and cells were washed with ice cold PBS and scraped in 180 µl RIPA lysis buffer (50mM Tris-HCl, pH 7.5, 200 mM NaCl, 1% Triton, 0.25% deoxycholic acid, 1 mM EDTA, 1mM EGTA) containing EDTA-free protease inhibitors (#11873580001, Roche) and phosphatase inhibitors (#78420, Pierce). Then collected cells were lysed mechanically and rotated for 1 h at 4°C. To remove nuclei, the samples were spun down (10’000 rpm, 10 min) and the protein concentration of the supernatant was determined by DC Protein Assay (Bio-Rad). 20 - 80 µg of protein was complemented with 6x sample buffer (200 mM Tris-HCl pH 6.8, 26% glycerol, 10% SDS, 0.01% bromphenol blue) and DTT (final concentration of 100 mM) and heated for 10 min at 95°C. Protein samples were loaded on 7-12% gels and SDS-PAGE was performed at 150 V. Transfer to nitrocellulose membranes (Protran BA83, Whatman Schleicher und Schuell) was applied at 100 V in the cold room for 1 hour and the blots were blocked for 2 hours with 5% milk powder in TBS-Tween (50 mM Tris HCl pH

7.4, 150 mM NaCl, 0.02% Tween). Primary antibodies were applied overnight and the secondary antibodies for 1 hour in 5% milk in TBS-Tween. The immunoblots were detected by enhanced chemiluminescence (#34094, Pierce) on Kodak BioMax light films (#Z370398- 50EA, Sigma). The list of primary and secondary antibodies employed with respective dilutions is detailed in table 3.

Antigen Species Conjugate Supplier (catalog number) Dilution

ACC1/2 Rabbit Purified Cell signaling (#3676) 1:500

AMPK Rabbit Purified Cell signaling (#2532) 1:1000

BiP Rabbit Purified Cell signaling (#3177) 1:500

CHOP Mouse Purified Santa Cruz (#sc-7351) 1:200

Mouse IgG Rabbit antiserum HRP Dako (#P0260) 1:4000

pAMPK Rabbit Purified Cell signaling (#2531) 1:200

pACC Rabbit Purified Cell signaling (#3661) 1:500

Rabbit IgG Goat antiserum HRP Dako (#P0448) 1:1600

ß-actin Mouse Purified Sigma (#A5441) 1:50,000

Table 4: List of primary and secondary antibodies for western blot.

2.7.

For measuring the β-oxidation of palmitic acid, tritium labeled palmitic acid (³H-palmitic acid, #NET043001MC Perkin Elmer, Schwerzenbach, Switzerland) was employed. For these experiments, Aicar was pre-incubated for 1h if necessary and serum starvation medium was employed having 0.2% FBS, 5 mM glucose which is supplemented with 0.5% FFA-free BSA.

For all the experiments 200µM of palmitic acid and 0.5µCi/ml ³H-palmitic acid was applied.

After the incubation times of experiment, 1ml of the supernatant medium was taken and added to 5ml of chloroform/methanol/5N HCl (2:1:0.05, v/v) and rotated for 5min. Later, the mixture was spun down at 350xg for 5 min, which apparently separates upper aqueous and lower organic phase. Now 500µl of upper aqueous phase was taken and added to 2ml of scintillation liquid (Insta-gel Plus, Packard, Groningen, The Netherlands) in a special scintillation reading tubes (Perkin elmer). Tubes were thoroughly mixed before they were put in scintillation reader (Packard, Canberra, CT, USA) for measuring radioactivity. β-oxidation values were obtained as disintegrations per minute (DPMI) and were normalized to total protein.

2.8. Statistical analysis

All the experiments were performed at least 4 times and representative result was shown.

Data are expressed as means ± SD unless otherwise mentioned. One way ANOVA was performed and for calculating significance of differences, Bonferroni post hoc test was employed. The prism 6 program was used for the analysis and differences were considered significant when P value was < 0.05.

3. RESULTS

3.1. Palmitic acid induced podocyte death: Modification by high glucose, TGF-β, and TNF-α

As reported by us and confirmed in independent studies, podocytes are highly susceptible to palmitic acid induced cell death (Sieber et al., 2010) (Sieber et al., 2013; Tao et al., 2012).

Previously other factors of the diabetic milieu have been reported to induce podocyte death, including high glucose concentrations (Susztak et al., 2006), TGF-β (Schiffer et al., 2001), and TNF-α (Ryu et al., 2012; Tejada et al., 2008). I explored the effect of these factors in combination with palmitic acid in podocytes.

3.1.1. Palmitic acid uncovers the toxicity of high glucose concentrations  

High glucose concentrations (20 – 30 mmolar) have been reported to induce apoptosis in podocytes (Susztak et al., 2006). In these studies apoptosis was quantified by assessment of nuclear condensation following DAPI staining and by a caspase 3 activity assay (Susztak et al., 2006). To see whether high glucose in combination with palmitic acid affects podocyte survival, I treated podocytes with normal glucose (NG, 11mM) or high glucose (HG, 22mM) in the presence or absence of 200µM palmitic acid for 48h (Figure 9). Podocyte cell death was assessed by flow cytometry after staining with Annexin V and propidium iodide (PI).

Annexin V-positive/PI-negative podocytes were considered apoptotic, whereas annexin V- positive/PI-positive podocytes were considered (late apoptotic) necrotic cells (Sieber et al., 2010). As shown in figure 9, high glucose alone had no significant effect on podocyte death in the presence of BSA which was used as the appropriate control for podocytes treated with palmitic acid complexed to BSA. However, high glucose significantly increased podocyte apoptosis in the presence of palmitic acid (155.3 ± 16.8%, p < 0.01). As the toxicity of high glucose was only seen in the presence of palmitic acid in these experiments we can also say that toxicity of glucose was “uncovered” by palmitic acid.

Figure 9: High glucose accentuates the to toxicity caused by palmitic acid in podocytes. Podocytes were incubated with either 11mM glucose (NG) or 22mM glucose (HG) in presence of 200µM palmitic acid or BSA (control) for 48h. 11mM of mannitol was employed to NG conditions to correct for osmolality. Bar graph represents mean percentages ± SD of apoptotic and necrotic cells (n = 3, * *p < 0.01).

3.1.2. TGF-β aggravates palmitic acid induced podocyte death  

Considerable evidence suggests that TGF-β plays an important role in the pathogenesis of DN (Chen et al., 2003), and TGF- β has been reported to induce apoptosis in murine podocyte (Schiffer et al., 2001). Therefore podocytes were treated with either 200µM palmitic acid or BSA (control) in the presence or absence of 5ng/ml TGF-β for 48h (figure 10). TGF-β could accentuate the toxicity caused by palmitic acid, as both apoptosis and necrosis levels were increased by 157.4 ± 10.2% (p<0.01) and 145.5 ± 10.6% (p<0.01) respectively. The effect of TGF-β alone on podocyte death was minimal and did not reach statistical significance, however the toxicity of TGF-β was again uncovered by the co-treatment with palmitic acid.

Figure 10: TGF TGFβ increases the toxicity caused by palmitic acid in podocytes: Podocytes were treated either with 200µM palmitic acid or BSA (control) with or without 5ng/ml TGFβ for 48h. Bar graph represents mean percentages ± SD of apoptotic and necrotic cells (n = 3, * *p < 0.01).

0 2 4 6 8 10 12 14 16 18

NG+BSA HG+BSA NG+palm HG+palm

Apoptosis   Necrosis  

% apoptotic/necrotic cells

NS        **  

0 5 10 15 20 25

BSA TGFβ+BSA palm TGFβ+palm

Apoptosis   Necrosis  

% apoptotic/necrotic cells

**

**

3.1.3. TNF alpha aggravates palmitic acid induced podocyte death  

A variety of direct and indirect evidence suggests that tumor necrosis factor alpha (TNF-α) plays a role in the pathogenesis of DN (Navarro-Gonzalez et al., 2009). Also, TNF-α was reported to induce podocyte death (Ryu et al., 2012; Tejada et al., 2008). To examine whether TNF-α affects palmitic acid induced podocyte death, podocytes were treated with 5ng/ml of TNFα alone or in combination with 200µM palmitic acid for 48h (Figure 11). TNF-α significantly increased podocyte death, specifically apoptosis was increased by 168.1 ± 20.9%

(p < 0.01), and necrosis by 120.6 ± 10.2% (p < 0.05). However, the increase of palmitic acid induce podocyte death by TNF-α was modest, and only apoptosis was significantly increased by 116.0 ± 7.9 (p < 0.05).

Figure 11: TNFα induces podocyte death and aggravates palmitic acid induced toxicity. Podocytes were treated with either 200 µM BSA or palmitic acid in presence or absence of 5 ng/ml TNFα for 48 h. Bar graph represents mean percentages +/- SD of apoptosic necrotic cells (n=3, **p < 0.01, *p < 0.05).

0 2 4 6 8 10 12 14 16 18

BSA TNFα+BSA palm TNFα+palm

Apoptosis Necrosis

% apoptotic/necrotic cells

     NS    *      

       *    

 **  

3.2. Regulation of fatty acid oxidation in palmitic acid induced podocyte cell death: Critical role of Acetyl CoA carboxylase 1 and 2

3.2.1. Differential regulation of genes involved in fatty acid metabolism in glomeruli of patients with established DN

To see whether enzymes of fatty acid metabolism might be regulated in DN, we performed microarray analysis of different enzymes involved in fatty acid metabolism in glomeruli from type 2 diabetic patients with DN and compared them to glomeruli from pretransplantation living donors (Sieber et al., 2013). Significantly altered expression levels of several enzymes involved in FAO and TG synthesis was observed (Figure 12). The most prominent change was the induction of SCD-1, which provides DGATs with their preferential substrates, MUFAs. Together with the positive regulation of DGAT1, which catalyzes the incorporation of exogenous FFAs into TG, this implies a disposition towards increased TG synthesis.

Furthermore, we saw a significant upregulation of all three isoforms of CPT-1, the rate- limiting enzyme for fatty acid oxidation, and a downregulation of ACC-2, which catalyzes the formation of the CPT-1 inhibitor malonyl-CoA, which suggests a disposition for increased fatty acid oxidation.

Figure 12: Differential expression of genes related to fatty acid metabolism in glomeruli of DN patients.

Microarray data were obtained from isolated glomeruli of type 2 diabetic patients with DN and controls (pretransplant allograft biopsies). Expression of fatty acid oxidation related genes such as ACC2, CPT-1a, CPT- 1b and CPT-1c, were significantly regulated in DN compared to controls. Up regulated enzymes are indicated in red, down regulated enzymes in blue colors.

3.2.2. Modulation of fatty acid oxidation and its effect on palmitic acid induced podocyte death

3.2.2.1. AMPK activation protects from palmitic acid induced cytotoxicity

To investigate whether stimulation of fatty acid oxidation (FAO) plays a protective role in palmitic acid treated podocytes, we took advantage of the AMP-activated protein kinase (AMPK) activator 5-aminoimidazole-4-carboxamide-1β-D-ribofuranoside (Aicar). Aicar (as

well as adiponectin (Sharma et al., 2008)) acts by phosphorylating AMPK, which in turn phosphorylates and inhibits ACC resulting in disinhibition of CPT-1 (Figure 13).

Figure 13: Scheme representing the metabolic path activated by Aicar and Adiponectin. Aicar and Adiponectin activate AMPK by phosphorylation. AMPK phosphorylates and inhibits ACC which results in decreased synthesis of malonyl CoA and disinhibition of CPT-1 resulting in upregulation of fatty acid oxidation.

In a first step, phosphorylation of AMPK and ACC by Aicar in podocytes was examined by Western immunoblotting (Figure 14).

Figure 14: Aicar phosphorylates AMPK and ACC. Immunoblot shows phosphorylation of AMPK and ACC after incubation of podocytes with either (PBS) vehicle or 0.5 mM Aicar for 14 hours. Total AMPK and total ACC served as loading controls.

Functionally, as shown in Figure 15, Aicar significantly prevented palmitic acid induced podocyte death assessed by flow cytometry after staining for Annexin V and propidium

iodide (PI). Specifically, Aicar reduced palmitic acid induced apoptosis and necrosis by 50.5

± 1.5% (p < 0.01) and 42.5 ± 6.1% (p < 0.05) respectively.

Figure 15: Aicar attenuates palmitic acid induced podocyte death. Podocytes were treated with either 200µM palmitic acid or BSA (control) with or without 0.5mM Aicar for 48h. Podocytes were preincubated with Aicar for 1h. Representative bar graph shows mean percentages ± SD apoptotic and necrotic podocytes. (n=3, *p<0.05,

**p<0.01)

Similarly to Aicar, the physiological AMPK agonist adiponectin (Sharma et al., 2008) also reduced palmitic acid induced podocyte death, although to a lesser extent than Aicar (figure 16). Specifically, adiponectin significantly decreased apoptosis by 14.1 ± 4.7% (p < 0.05), but the reduction of necrosis by 9.9 ± 6.3% did not reach statistical significance. To see the protective effect of adiponectin podocytes were kept at a high glucose concentration of 22 mmol/L, which is known to reduce phosphorylation of AMPK (Sharma et al., 2008) and increases the susceptibility of podocytes to AMPK activation (Sharma et al., 2008).

Figure 16: Adiponectin decreases apoptosis induced by palmitic acid. Podocytes were treated with either 200µM palmitic acid or BSA (control) with or without 0.5µg/ml adiponectin for 48h. 22mM of glucose was employed.

0 2 4 6 8 10 12 14

BSA BSA

0.5mM Aicar

palm palm 0.5mM Aicar

Apoptosis Necrosis

   **  

*  

  %  apoptoZc/necroZc  cells      **  

 **  

0   2   4   6   8   10   12   14   16   18  

BSA BSA+Adipo Palm Palm+Adipo

Apoptosis Necrosis

%  apoptoZc/necroZc  cells  

**    

**  

     *                            NS  

Podocytes were preincubated with adiponectin for 1h. Bar graph represents mean percentages ± SD of apoptotic and necrotic cells (n = 3, ** p < 0.01, * p < 0.05).

3.2.2.2. Inhibition of AMPK exacerbates palmitic acid induced cell death and reversed the protection caused by Aicar.

To further to explore the role of AMPK we used the AMPK inhibitor compound C.

Compound C was used at a low concentration of 4 µM, as higher concentrations were toxic, i.e. podocyte death was markedly increased for BSA control (data not shown). The Aicar induced ACC phosphorylation was significantly reduced by compound C (Figures 17A, and 17B, p < 0.05).

A)

B)

In line with the inhibitory effect of compound C on the AMPK-ACC-CPT-1 pathway, compound C treatement increased palmitic acid induced apoptosis and necrosis in podocytes by 140.1 ± 20.1 % (p < 0.01) and 130.9 ± 14.0% (p < 0.01), respectively. In agreement with the partial reduction of the Aicar induced ACC phosphorylation, (Figure 17) cotreatment with compound C compared to Aicar alone only moderately increased palmitic acid induced podocyte death, i.e. apoptosis was increase by 128.2 ± 9.3 % (NS) and necrosis by 176.7 ± 9.7% (p < 0.01) (Figure 18).

0 200 400 600 800

DMSO Aicar Aicar+CompC

pACC

pACC expression (% Rel. to ACC) *

Figure 17: Compound C inhibits Aicar induced phosphorylation of ACC. A.

Podocytes were incubated with DMSO, 0.5mM Aicar and Aicar in combination of 4µM of compound C. pACC was immunoblotted and total ACC served as loading control. B. Quantification of pACC by densitometry. Bar graph represents the relative expression ± SD (*p < 0.05).

DMSO treated controls were set to 100%.